Selective Photoelectrochemical Detection of DNA with High-Affinity

Department of Biological Sciences and Biotechnology, Medical Systems Biology Research Center, Department of Chemistry, and State Key Laboratory of ...
0 downloads 0 Views 88KB Size
Download PDF
Anal. Chem. 2006, 78, 4722-4726

Correspondence

Selective Photoelectrochemical Detection of DNA with High-Affinity Metallointercalator and Tin Oxide Nanoparticle Electrode Shili Liu,†,‡ Chao Li,§ Jing Cheng,†,‡,|,⊥ and Yuxiang Zhou*,†,‡,|,⊥

Department of Biological Sciences and Biotechnology, Medical Systems Biology Research Center, Department of Chemistry, and State Key Laboratory of Biomembrane and Membrane Biotechnology, Tsinghua University, Beijing 100084, China, and National Engineering Research Center for Beijing Biochip Technology, 18 Life Science Parkway, Changping District, Beijing 102206, China

Selective detection of double-stranded DNA (ds-DNA) in solution was achieved by photoelectrochemistry using a high-affinity DNA intercalator, Ru(bpy)2dppz (bpy ) 2,2′bipyridine, dppz ) dipyrido[3,2-a:2′,3′-c]phenazine) as the signal indicator and tin oxide nanoparticle as electrode material. When Ru(bpy)2dppz alone was irradiated with 470-nm light, anodic photocurrent was detected on the semiconductor electrode due to electron injection from its excited state into the conduction band of the electrode. The current was sustained in the presence of oxalate in solution, which acted as a sacrificial electron donor to regenerate the ground-state metal complex. After addition of double-stranded calf thymus DNA into the solution, photocurrent dropped substantially. The drop was attributed to the intercalation of Ru(bpy)2dppz into DNA and, consequently, the reduced mass diffusion of the indicator to the electrode, as well as electrostatic repulsion between oxalate anion and negative charges on DNA. The degree of signal reduction was a function of the DNA concentration, thus forming the basis for real-time DNA detection. The signal reduction was selective for ds-DNA, as no such effect was observed for single-stranded polynucleotides such as poly-G, poly-C, poly-A, and poly-U. The detection limit of calf thymus ds-DNA reached 1.8 × 10-10 M in solution. DNA detection and quantification have become increasingly important as scientists unravel the genetic basis of disease and use this new information to improve medical diagnosis and treatment. Methods that are rapid and easy to detect and quantify nucleic acids are preferable. Hybridization of nucleic acids to their complementary sequences is the essence of modern molecular * Corresponding author. Phone: 86-10-80726786. Fax: 86-10-80726896. Email: [email protected]. † Department of Biological Sciences and Biotechnology, Tsinghua University. ‡ Medical Systems Biology Research Center, Tsinghua University. § Department of Chemistry, Tsinghua University. | State Key Laboratory of Biomembrane and Membrane Biotechnology, Tsinghua University. ⊥ National Engineering Research Center for Beijing Biochip Technology.

4722 Analytical Chemistry, Vol. 78, No. 13, July 1, 2006

biology. The majority of the current methods for the detection of nucleic acid sequence are coupled with sophisticated fluorescent detection technology.1 However, the fluorescence-based method suffers from the drawback of high cost due to the complex and expensive optical detection instrumentation. Nonoptical methods have been developed and reported.2 Photoelectrochemistry is an analytical method that has great potential but is currently underexploited. In the method, an electron of a photoelectrochemically active molecule is excited from its ground state to the excited state after absorbing photoenergy. If an electrode with appropriate energy level is in the vicinity, the electron transfers to the electrode and generates photocurrent. To carry out photoelectrochemical detection, one would use light as the excitation source and detect the photocurrent. Therefore, the sensitivity of this method is potentially very high due to the separation of excitation source and detection signal. An additional benefit of photoelectrochemistry in comparison with all optical detection methods is its low cost inherent to electronic detection, which becomes a significant issue in arraybased analysis. The principle of photoelectrochemistry has been applied to solar energy conversion for decades. The most successful system in terms of quantum yield was developed by Gratzel and coworkers,3 which is composed of a thick, porous film of nanocrystalline TiO2 particles sensitized by surface-adsorbed ruthenium polypyridyl complex. Other wide band-gap semiconductors and sensitizers have also been investigated.4-7 Given the potential of photoelectrochemistry as a sensitive analytical method, however, it is surprising that not much work has been done in this area. Weber et al. first reported photoelectrochemical detection of ruthenium tris(bipyridine) (Ru-bpy; bpy ) 2,2′-bipyridine) on a (1) Epstein, J. R.; Brian, I.; Walter, D. R. Anal. Chim. Acta 2002, 469, 3. (2) Kricka, L. J. Ligand-Binder Assays: Labels and Analytical Strategies; Marcel Dekker: New York, 1985. (3) O′Regan, B.; Gratzel, M. Nature 1991, 353, 737. (4) Kalyanasundaram, K.; Gratzel, M. Coord. Chem. Rev. 1998, 77, 347. (5) Kelly, C. A.; Meyer, G. J. Coord. Chem. Rev. 2001, 211, 295. (6) Kamat, P. V.; Bedja, I.; Hotchandani, S.; Patterson, L. K. J. Phys. Chem. 1996, 100, 4900. (7) Giraudeau, A.; Fan, F. R.; Bard, A. J. J. Am. Chem. Soc. 1980, 102, 5137. 10.1021/ac052022f CCC: $33.50

© 2006 American Chemical Society Published on Web 06/03/2006

glassy carbon electrode.8 Upon photoexcitation, the ruthenium complex transfers an electron to a quencher molecule in solution. The oxidized complex is subsequently reduced at the electrode, resulting in photocurrent generation. The idea of using Ru-bpy as a photoelectrochemical label in chemical and biological analysis was proposed by the authors, but no experiments were done. In a first attempt to detect DNA in solution by photoelectrochemistry, Pandey and Weetall9 employed a DNA-intercalating organic dye, anthraquinone, as the indicator. The signal-generating mechanism is similar to the one described by Weber et al. in that photoexcited anthraquinone is reduced by an electron donor in solution. The reduced, ground-state indicator is then detected at a modified graphite carbon electrode at an oxidizing potential. Transition metal complexes with dppz (dppz ) dipyrido[3,2-a:2′,3′-c]phenazine) and other planar ligands have been found10 to bind with double-stranded (ds)-DNA with high affinity (K ) 106-107 M-1). Taking this advantage, hybridized oligonucleotides on a gold electrode surface were detected photoelectrochemically with highaffinity metallointercalators.11,12 The mechanism of photocurrent generation was not clear, although the authors proposed an electron-transfer process from the photoexcited Ru(II) to the gold electrode and subsequent reduction of Ru(III) by a reducing agent in solution. Most recently, using covalently linked anthraquinone as photoelectrochemical label, Okamoto et al. investigated charge transport through DNA duplex immobilized on a gold electrode and observed DNA sequence-dependent photocurrent.13 DNA detection using semiconductor nanoparticles has also been reported.14 We have successfully applied the principle of dye-sensitized photoelectrochemistry to the quantitative detection of biological affinity reactions.15 Using a Ru-bpy derivative as photoelectrochemical label, oxalate as the sacrificial electron donor, and nanoparticle SnO2 as electrode material, binding of bovine serum albumin with surface-immobilized avidin through biotin was detected in the range of 1-100 µg/mL albumin. This analytical system possesses many attributes of the high quantum yield solar cell, such as high electrode surface area, optimal match of energy level between label and SnO2 conduction band edge, fast charge separation on semiconductor electrode, and recycling of the label. Therefore, it can potentially perform better in photoelectrochemical detection than the previous systems described above, where conducting electrodes such as gold and carbon were used. This report describes the application of our photoelectrochemical system in the detection of DNA in solution with Ru(bpy)2dppz, a high-affinity DNA intercalator, as the signal generator. Its photoelectrochemical current was amplified by a sacrificial electron donor, oxalate. Similar to the observations reported by Carter and Bard16 and Li et al.17 in their voltammetric measurement, (8) (9) (10) (11) (12) (13) (14) (15) (16)

Weber, S. G.; Morgan, D. M.; Elbicki, J. M. Clin. Chem. 1983, 29, 1665. Pandey, P. C.; Weetall, H. H. Anal. Chem. 1994, 66, 1236. Erkkila, K. E.; Odom, D. T.; Barton, J. K. Chem. Rev. 1999, 99, 2777. Nakamura, S.; Shibata, A.; Takenaka, S.; Takagi, M. Anal. Sci. 2001, 17 (Suppl), 1431. Gao, Z.; Tansil, N. C. Nucleic Acids Res. 2005, 33, e123. Okamoto, A.; Kamei, T.; Tanaka, K.; Saito, I. J. Am. Chem. Soc. 2004, 126, 14732. Willner, I.; Patolsky, F.; Wasserman, J. Angew. Chem., Int. Ed. 2001, 40, 1861. Dong, D.; Zheng D.; Wang F.-Q.; Yang X.-Q.; Wang N.; Li, Y.-G.; Guo, L.H..; Cheng, J. Anal. Chem. 2004, 76, 499. Carter, M. T.; Bard, A. J. J. Am. Chem. Soc. 1987, 109, 7528.

intercalation of the metal complex into DNA led to a reduction of the photogenerated current due to slower mass diffusion of the indicator and electrostatic repulsion between oxalate anion and DNA phosphate groups. By the combination of high-affinity intercalator and amplified photoelectrochemical method, 1.8 × 10-10 M ds-DNA was detected in solution. EXPERIMENTAL SECTION Sodium oxalate was purchased from Avocado Research Chemicals (Lancaster, PA) and oxalic acid from VAS Lab Supplies (Tianjin, PR China); Ru(bpy)2dppz was synthesized according to the published procedure.18 Calf thymus ds-DNA (13K base pairs), polygunydic acid (poly-G), polycystilic acid (poly-C), polyadenic acid (poly-C) and polyutandilic acid (poly-U) were obtained from Sigma Chemical Co (St. Louis, MO) and used as received. According to the manufacturer, the four single-strand polynucleotides are a mixture of nucleic acids with a wide range of molecular sizes; therefore, the chain length or number of bases is not specified. Four oligonucleotides for PCR and hybridization were purchased from Invitrogen (Carlsbad, CA), which are composed of the following base sequences: 5′-TTCAGCGGGGAGGAAGGGAGTAAAGTTAATACCTTTGCTCATTGACGTTACCCGC-3′ (T1);5′-GCGGGTAACGTCAATGAGCAAAGGTATTAACTTTACTCCCTTCCTCCCCGCTGAA-3′(T2,complementarytoT1);5′-AGAGTTTGATCCTGGCTCAG-3′ (forward primer); 5′-AAGGAGGTGATCCAGCC-3′ (reverse primer). In a typical symmetric PCR experiment, the mixture contained 5 µL of 10× PCR buffer solution (200 mM Tris-HCl, pH 8.4, 500 mM KCl), 4 µL of 25 mM MgCl2, 1 µL of 10 mM dNTPs, 2 µL of 10 µM forward primer, 2 µL of 10 µM reverse primer, 0.5 µL of 1 ng/µL DNA template, 0.5 µL of 5 units/µL Taq DNA polymerase, and 35 µL of autoclaved double-deionized water. The DNA template was Staphylococcus aureus genomic DNA. The Mastermix was subjected to the following thermal cycling profile: initial denaturation at 94 °C for 10 min, 40 cycles at 94 °C for 1 s, at 59 °C for 1 min, at 72 °C for 1 min, and a final extension at 72 °C for 5 min. T1 and T2 were dissolved in pH 8.0 Tris-HCl solution and hybridized at room temperature for 1 day. PCR product (ds-PCR DNA) and T1-T2 hybridization product (ds-oligo) were electrophoresed, harvested with DNA Gel Extraction Kit (V-gene, Beijing, PR China), and precipitated in ethanol. Colloidal SnO2 and its thick film on conductive glass were prepared by following the previously described procedure.15,19 Photocurrent was measured on a CH Instrument model 800 electrochemical analyzer (CH Instrument, Inc.) using a Pt flag counter electrode, a Ag/AgCl reference electrode, and a bias voltage of +0.3 V. The excitation light source was a blue lightemitting diode (Lamp Inc., Shenzhen, PR China) with an illumination area of 0.2 cm2. In the action spectrum measurement, a 500-W Xe lamp and a monochromator were employed to provide variablewavelength excitation. (17) Li, C.; Liu, S. L.; Guo, L. H.; Chen, D. P. Electrochem. Commun. 2004, 7, 23. (18) Musumeci, S.; Rizzarelli, F. S.; Sammartano, S.; Bonomo, R. P. Inorg. Chim. Acta 1973, 7, 660. (19) Mulvaney, P.; Grieser, F.; Meisel, D. Langmuir 1990, 6, 567.

Analytical Chemistry, Vol. 78, No. 13, July 1, 2006

4723

Figure 1. Time-based photocurrent response of SnO2 nanoparticle electrode in (a) 100 mM phosphate, pH 5.8; (b) 100 mM oxalate/100 mM phosphate, pH 5.8; (c) 10 µM Ru(bpy)2dppz, 100 mM phosphate, pH 5.8; and (d) 10 µM Ru(bpy)2dppz, 100 mM phosphate/100 mM oxalate buffer, pH 5.8. Light excitation of 470 nm was switched every 10 s.

RESULTS AND DISCUSSION Interaction of metal complexes with DNA in solution has been studied by Bard’s and Thorp’s group by cyclic voltammetry.16,20,21 Change of the voltammograms was observed after the metal complex binding with DNA. Basically, voltammetric peak current decreased due to slower mass diffusion of the much larger DNA/ metal complex. The attenuation could be used in the detection and quantification of DNA in solution. Ru(bpy)2dppz was selected as a DNA intercalating indicator for two reasons. One is that dppz ligand has the highest binding constant to date,10 with K ) 106107 M-1. The other reason is that the photoelectrochemical property of Ru(bpy)2dppz is similar to Ru(bpy)32+, which has already been used as a photoelectrochemical label in our previous work for the quantitative analysis of biotin-avidin affinity reaction. Recently, this complex was used by Li et al. for the voltammetric determination of double-stranded DNA in solution.17 By combination with oxalate in a catalytic reaction, high detection sensitivity was achieved. In principle, even higher sensitivity can be obtained in the photoelectrochemical detection using Ru(bpy)2dppz as the indicator. This is because photoelectrochemistry has an inherent advantage over voltammetry in having separate signals for excitation and detection, very much like the highly sensitive electrochemiluminescence technique. This is the major motivation of the current work. Chemical amplification mechanism in electrochemical detection can improve the signal-to-background ratio. Our previous study has shown that the oxidation current of Ru-bpy was amplified by more than 1000-fold in an oxalate solution.22 However, the comparison has not been done in photoelectrochemisty. We measured the photocurrent of both oxalate-free and oxalatecontaining electrolytes, and the i-t curve is illustrated in Figure 1. As shown very clearly, when there is no Ru(bpy)2dppz in solution, photocurrent is marginally distinguishable from the dark current. In contrast, in the presence of 10 µM Ru(bpy)2dppz, a substantial rise of current is clearly visible when the excitation light in turned on, which proves that the response observed on (20) Rodriguez, M.; Bard, A. J. Anal. Chem. 1990, 62, 2658. (21) Welch, T. W.; Corbett, A. H.; Thorp, H. H. J. Phys. Chem. 1995, 99, 11757. (22) Zheng, D.; Wang, N.; Wang, F. Q.; Dong, D.; Li, Y. G.; Yang, X. Q.; Cheng, J.; Guo, L. H. Anal. Chim. Acta 2004, 508, 225.

4724 Analytical Chemistry, Vol. 78, No. 13, July 1, 2006

Figure 2. Photocurrent response of SnO2 nanoparticle electrode as a function of excitation wavelength in 100 mM oxalate/100 mM phosphate, pH 5.8 (9) and 50 µM Ru(bpy)2dppz in 100 mM oxalate/ 100 mM phosphate, pH 5.8 (b). Inset: absorption spectrum of Ru(bpy)2dppz.

the electrode is due to the photoexcitation of Ru(bpy)2dppz. The direction of the current is anodic, suggesting that the current is produced by electron transfer of the excited-state metal complex to the electrode. In the absence of oxalate, the photocurrent was sustained by the reaction between Ru(bpy)2dppz(III) and water.23 Comparing oxalate-free and oxalate-containing solution, the current of the latter is twice as much, leading to higher signal/ background ratio (here background is defined as the photocurrent in the absence of the metal complex). The signal was found to be proportional to the illumination power up to the maximum of the light-emitting diode (329 µW) and was quite stable over the 60-s time span. Figure 2 shows the photocurrent response as a function of excitation wavelength. In a solution containing Ru(bpy)2dppz, the photocurrent response changed with the excitation wavelength, and reached its peak at 470 nm. In solutions without the metal complex, the photocurrent is low regardless of the excitation wavelength. The photocurrent action spectrum of Ru-dppz has a shape similar to the UV-visible absorption band of Ru-dppz, providing another piece of evidence that the photocurrent is associated with Ru-dppz excitation. Based on the experimental result, we chose 470 nm as the excitation wavelength in the following experiments. After addition of calf thymus DNA into a Ru(bpy)2dppz solution, photocurrent reduced substantially. A similar effect was reported by Pandey and Weetall with anthraquinone as the photoelectrochemical indicator.9 There are at least three possible reasons for the reduction of photocurrent: (a) impeded electrode reaction of Ru-dppz after intercalating into DNA; (b) reduced mass diffusion due to the increased size of the DNA/Ru-dppz adduct; and (c) electrostatic repulsion between DNA phosphates and oxalate anion, which interferes with the catalytic reaction between intercalated Ru-dppz and oxalate. Based on the cyclic voltammetric data provided in the literature,16,20,21 it is believed the heterogeneous electron-transfer rate of the metal complex was not affected after DNA intercalation. The last possibility does not exist in the regular voltammetry employed by Bard and Thorp. Nor does it apply to Pandey and Weetall’s case where a neutrally charged electron donor, glucose, was used. The electrostatic effect was investigated in detail in our previous study on the voltammetric investigation of DNA/Ru-dppz binding. Therefore, we attribute (23) Ghosh, P. K.; Brunschwig, B. S,; Chou, M.; Creutz, C.; Sutin, N. J. Am. Chem. Soc. 1984, 106, 4772.

Figure 4. Relative photocurrent response as a function of ds calf thymus DNA concentration in the range of 0.018 and 18 nM. Photocurrent measurement was performed in 10 µM Ru(bpy)2dppz/ 30 mM oxalate buffer, pH 5.8 with 470-nm light excitation.

Figure 3. Relative steady-state photocurrent response as a function of (a) the molar concentration, and (b) mass concentration of ds calf thymus DNA (0), ds-PCR DNA (b), ds-oligo DNA ([), poly-G (1), poly-C (+), poly-A (left solid triangle), poly-U (right solid triangle), oligo-C (2), and ss-oligo DNA (-). Photocurrent measurement was performed in 10 µM Ru(bpy)2dppz/30 mM oxalate buffer, pH 5.8 with 470-nm light excitation.

the photocurrent decrease to the intercalation of Ru-dppz into DNA and consequently the reduced mass diffusion of the indicator to the electrode, as well as electrostatic repulsion between oxalate and negative charges on DNA. The diffusion coefficient of Os(bpy)2dppz was found by Welch et al to be ∼20 times less when bound to calf thymus DNA.21 Due to the similar structure of the Os and Ru complexes, the same reduction in diffusion coefficient for DNA-bound Ru(bpy)2dppz is anticipated. The degree of photocurrent reduction depends on the concentration ratio of Ru(bpy)2dppz/DNA. Obviously, a low indicator concentration would bring about high sensitivity for DNA detection. But if the indicator concentration is too low, S/B ratio would not be enough. After optimization, Ru(bpy)2dppz concentration was chosen as 10 µM. Figure 3 illustrates the photocurrent response as a function of DNA concentration. The degree of signal reduction was found to be a function of ds-DNA concentration, thus forming the basis for real-time DNA detection. The detection was selective for ds-DNA, as no such effect was observed for polynucleotides such as poly-G, poly-C, poly-A, and poly-U. The photocurrent for the four single-strand polynucleotides remained essentially unchanged in the concentration range examined, suggesting that under the experimental conditions the electrostatic attraction between Ru(bpy)2dppz2+ and the phosphate groups on

nucleotides is not strong enough to change its mass diffusion. Figure 3 also shows that, at the same molar concentration, the change of photocurrent depends strongly on the DNA chain length. The signal reduction is most pronounced with calf thymus DNA (13K base pairs), less with the PCR product (1.5K base pairs), and almost none with the hybridized oligonucleotide (55 base pairs). This is because, at the same molar concentration, shorter DNAs have less intercalation sites for Ru(bpy)2dppz to bind to. When the mass concentration is used instead, a similar change in signal is observed regardless of DNA length. The signal loss associated with the single-stranded oligonucleotide is most likely due to the intrastrand base pairing and indicator intercalation to these base pairs. According to our previous work,17 a buffer of 30 mM oxalate/ oxalic acid, pH 5.8, balances the need for high oxalate concentration and low ionic strength. We therefore used this buffer to find the lower detection limit in the photoelectrochemistry measurement. As shown in Figure 4, the detection limit for double-stranded calf thymus DNA is ∼0.18 nM, with a dynamic range from 0.5 to 7 nM. Although no specific number was given in Pandey and Weetall’s report9 for the detection limit, the lowest DNA concentration in Figure 6 of their report is 4 µg/20 µL, or 300 nM, which is 3 orders of magnitude inferior than the detection limit of this work. The major difference of the two photoelectrochemical systems lies in the indicator (Ru-dppz vs anthraquinone) and electrode material (SnO2 vs carbon). The high sensitivity obtained in this work demonstrates the superiority of the new system, which gets its components from the well-developed and highly optimized dye-sensitized photoelectrochemical cell by using a high-affinity metallointercalator, a porous semiconductor electrode, and an electron donor. Although the sensitivity for PCR products and oligonucleotides is rather limited, the sensitivity with long DNA is adequate for some applications. One example is the quantification of total DNA in organisms that are usually more than 10 kb long, such as DNA extracted from the microorganisms in soil. There is also a need for the quantification of long DNA in biological research. One such example is the artificial chromosome (BAC, YAC, etc.), where the insert fragment is quite large (the average insert length of BAC is 120 kb). In addition, PCR is heading for the amplification of larger fragment DNA (>5 kb), which is significant in gene research and genetic disease detection.24-27 There are already (24) Barnes, W. M. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 2216.

Analytical Chemistry, Vol. 78, No. 13, July 1, 2006

4725

some commercial PCR kits for the amplification of large-fragment DNA (e.g., Takara RNA LA PCR TM Kit). For the detection of DNA hybridization of oligonucleotides, we need to work with a solid-phase system. Our current work has already demonstrated significantly improved sensitivity over previous methods on the detection of DNA in solution. We hope similar improvement can be made for the detection of solid-phase DNA hybridization. CONCLUSIONS A newly developed photoelectrochemical detection system was applied to the quantification of DNA in solution. The system employed a high-affinity DNA intercalator, Ru(bpy)2dppz, as the photoelectrochemical indicator, porous SnO2 film as semiconductor electrode, and oxalate as electron donor in solution to sustain (25) Cheng, S.; Fockler, C.; Barnesf, W. M.; Higuchi, R. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 5695. (26) Basso, K.; Frascella, E.; Zanesco, L.; Rosolen A. Am. J. Pathol. 1999, 155, 1479. (27) Lindberg, A. M.; Polacek, C.; Johansson, S. J. Virol. Methods 1997, 65, 191.

4726

Analytical Chemistry, Vol. 78, No. 13, July 1, 2006

photocurrent response. After addition of double-stranded calfthymus DNA into solution, the current dropped substantially. The effect was not observed in poly-G, poly-C, poly-A, and poly-U, illustrating the high selectivity of the indicator. The degree of signal reduction was a function of DNA concentration, thus forming the basis for DNA quantitation. A detection limit of 1.8 × 10-10 M was obtained, which is 3 orders of magnitude more sensitive than the system utilizing organic dye and conducting electrode. ACKNOWLEDGMENT We gratefully acknowledge the financial support from the National Hi-Tech Program of China (2002AA2Z2011).

Received for review November 14, 2005. Accepted April 12, 2006. AC052022F